Update README.md
Browse files
README.md
CHANGED
|
@@ -62,11 +62,9 @@ state_dict = load_file("model.safetensors")
|
|
| 62 |
|
| 63 |
## Limitations
|
| 64 |
|
| 65 |
-
- PiEa is an
|
| 66 |
- It is trained on image reconstruction and denoising, not prompt following.
|
| 67 |
- Quality depends on the surrounding pipeline and conditioning signal.
|
| 68 |
-
- The small checkpoint is optimized for throughput and experimentation rather than maximum fidelity.
|
| 69 |
-
- Evaluate on your target image distribution before use.
|
| 70 |
|
| 71 |
## Citation
|
| 72 |
|
|
|
|
| 62 |
|
| 63 |
## Limitations
|
| 64 |
|
| 65 |
+
- PiEa is an pixel decoder, not a full image generation system.
|
| 66 |
- It is trained on image reconstruction and denoising, not prompt following.
|
| 67 |
- Quality depends on the surrounding pipeline and conditioning signal.
|
|
|
|
|
|
|
| 68 |
|
| 69 |
## Citation
|
| 70 |
|